Every lidar point can have a classification code assigned to it that defines the type of object that has reflected the laser pulse. Lidar points can be classified into a number of categories, including bare earth or ground, top of canopy, and water. The classes are defined using numeric integer codes in the LAS files.
Lidar points stored in LAS files are typically classified into categories using specialized classification tools outside of ArcGIS Pro. This classification is usually completed by setting parameters based on the terrain and then running algorithms on the point cloud to determine the feature type associated with each point. The classification code assigned to each point is written to the LAS file and, in most cases, adheres to the American Society for Photogrammetry and Remote Sensing (ASPRS) standard.
When a classification is carried out on lidar data, points may fall into more than one category of the classification. Classification flags are used to provide a secondary description or classification for lidar points. In later versions (LAS 1.1 and later), class flags were used to solve this problem. Classification flags were added to the LAS standard to mark points with information that supplements the traditional classification. Synthetic, key-point, withheld, and overlap flags can be set for each lidar point. These flags can be set along with the classification codes. For example, a water record can be given a classification code for water (9) as well as a withheld flag. The point will remain in the dataset but will be withheld from any additional analysis on the LAS files.
In many cases, LAS files may not be fully or correctly classified when used as input for the GIS tools in ArcGIS Pro. ArcGIS Pro provides the LAS dataset and associated tools to enable classification or data cleanup of classification codes and classification flags residing in the LAS files.
The list below includes common examples of how you can benefit from editing LAS datasets:
- Manually fix errors in class codes.
- Visually compare the lidar points against existing GIS data for data validation, such as building data.
- Classify lidar points using GIS features.
- Reclassify lidar points by manually selecting one or many lidar points.
- Change all classification codes currently in the LAS dataset.
- Add or remove LAS files from a LAS dataset to increase or decrease the data area extent.
- Interactively measure 3D distances between visible features, such as power lines and trees.
- Use the lidar points as backdrop data to digitize new GIS features, such as street furniture.
- Measure height offsets and distances between points or overlapping datasets or flight lines.
- View high-accuracy control points alongside the LAS points and compare height differentials.
- View and compare two LAS datasets collected at different times for the same area of interest.
The interactive and geoprocessing tools in ArcGIS Pro provide the ability to edit classification codes present in LAS files.
Interactive LAS point classification
In many cases, LAS files may not be fully or correctly classified when used with ArcGIS Pro, and you may need to manually edit the classification of the LAS points. The Classification tab allows you to interactively manipulate the classification conducted on the LAS files of a LAS dataset in a scene. You can quickly change the class codes and classification flags that are currently set on the selected points.
The Classification tab for a LAS file or LAS dataset provides interactive editing capabilities for selected points from a 2D map or 3D scene view. You can select points directly in a 3D scene or optionally shift your scene into a side-on viewpoint using profile viewing. Profile viewing is a temporary display state that clips to a defined extent and is useful for vertically stacked data. Profile viewing is available in 3D scenes and not available from a 2D map. Profile viewing is available either from the View tab or directly from the Classification tab for a LAS dataset layer. In a profile viewing state, you can navigate or incrementally move through the points to more clearly identify, select, and reclassify the data. For more information on the navigation tools for a profile viewing state, see Adjust the profile view.
You can only edit the classification currently defined in the LAS files referenced by the LAS dataset. All changes are permanent when making any modifications to the classification codes in LAS files. If you are conducting any what-if scenarios or do not want changes to be permanent, make sure to work on a copy of the LAS files and not the originals. You can also use the Extract LAS tool to extract a portion of a LAS file to conduct an initial test. ZLAS files cannot be edited. Consider extracting them to LAS files and adding them again to a LAS dataset to edit the classification.
To interactively select and edit LAS points, do the following:
- Select a LAS dataset or LAS file layer in the Contents pane.
- Navigate to the location to select LAS points.
- Click the Classification tab.
- Optionally, use the tools in the Profile Viewing group to create a profile view of the extent of the points to be classified. For more information on profile viewing, see Profile viewing.
- Use the Select drop-down menu to choose a selection method. To learn more about selection, see Select features interactively.
- Select the points to be edited.
- From the Interactive Edit group, click the Classification Codes drop-down menu to choose the new classification code for the selected points. You can also set or clear classification flags. Select the classification flag and click either Set or Clear from the drop-down menu.
The classification codes and classification flags are described below.
- Click Apply Changes to commit the edits. All changes are permanent and cannot be undone.
- To exit profile viewing mode, click the Close button in the upper right of the scene view, in the same banner where the Profile Viewing text appears.
You can always click the Clear button to clear any selected points.
Emulate a map view to create a profile view
Profile viewing is only available in a 3D scene. If you prefer to use a 2D map view with your lidar data, follow these steps to create a similar 2D environment in a 3D scene:
- Navigate the camera position to look straight down in 3D. This allows you to pan around in a 3D view as you would in a 2D view.
- Click the View tab and click Drawing Mode . Set Drawing Mode to Parallel.
- Ensure that a ground elevation surface is turned on so that when you click in a blank space between points it is valid. The ground surface display may conflict with points, so set its color to be transparent.
- Select the Ground layer in the Contents pane.
- Click the Appearance tab.
- Click the Surface Color drop-down menu and choose No color.
- Use the tools in the Profile Viewing group to create a profile view of the extent of the points to be classified. For more information on profile viewing, see Profile viewing.
Geoprocessing tools for classification editing
Geoprocessing tools allow you to automate editing of classification codes.
Classifies unclassified LAS points into ground points.
|Classify LAS Building|
Classifies building rooftop points in aerial lidar data.
|Classify LAS By Height|
Reclassifies lidar points based on their height from the ground surface.
|Set LAS Class Codes Using Features|
Changes lidar classification codes assigned to lidar points based on their proximity to feature data. For example, breaklines representing water boundaries can be used to classify all points within the water boundaries as class code 9 to represent water.
|Locate LAS Points By Proximity||
Identifies lidar points within the three-dimensional proximity of z-enabled features while also providing the option to reclassify the points and export them to an output feature class.
|Change Class Codes||
Modifies the classification codes for LAS files referenced by a LAS dataset. This tool reclassifies one set of classification codes into another, which is particularly useful for updating the classification of LAS files generated before the introduction of classification standards in the LAS 1.1 specification. The tool can be used to ensure that such data conforms to the current standards.
Classifies LAS points from overlapping scans of aerial lidar surveys.
Classifies LAS points with anomalous spatial characteristics as noise.